Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference
-
Annu Int Conf IEEE Eng Med Biol Soc · Jul 2018
The Role of Baroreflex Sensitivity in Acute Hypotensive Episodes Prediction in the Intensive Care Unit.
A life threatening condition in Intensive Care Unit (ICU) is the Acute Hypotensive Episode (AHE). Patients experiencing an AHE may suffer from irreversible organ damage associated with increased mortality. Predicting the onset of AHE could be of pivotal importance to establish appropriate and timely interventions. ⋯ We will investigate the predictive power of features related to the baroreflex by performing classifications with and without them. Results are obtained using 17 classifiers belonging to different model families: classification trees, Support Vector Machines (SVMs), K-Nearest Neighbors (KNNs) replicated with different set of hyper-parameters and logistic regression. On average, the use of baroreflex features in the AHE prediction process increases the Area Under the Curve (AUC) by 10%.
-
Annu Int Conf IEEE Eng Med Biol Soc · Jul 2018
Improved Heart Rate Tracking Using Multiple Wrist-type Photoplethysmography during Physical Activities.
Photoplethysmography (PPG) signals collected from wearable sensing devices during physical exercise are easily corrupted by motion artifact (MA), which poses great challenge on heart rate (HR) estimation. This paper proposes a new framework to accurately estimate HR using two leads of PPG signals in combination with accelerometer (ACC) data in the presence of MA. A moving time window is first used to segment PPG signals and ACC signals. ⋯ The proposed method was validated using the 2015 IEEE Signal Processing Cup dataset. The average absolute error is 1.15 beats per minutes (BPM) (standard deviation: 2.00 BPM), and the average absolute error percentage is 0.95% (standard deviation: 1.86%). The proposed method outperforms the previously reported work in terms of accuracy.
-
Annu Int Conf IEEE Eng Med Biol Soc · Jul 2018
Selective Collection and Condensation of Exhaled Breath for Glucose Detection.
Exhaled breath condensate (EBC) is a promising non-invasive sample for the detection of various analytes, such as glucose. However, the methods used to collect EBC are highly inconsistent; the variable dilution factors associated with water vapor and the inclusion of dead space air significantly impact the reliability of reported analyte concentrations in EBC. For example, current EBC glucose measurements have resulted in dilution factors ranging from 1/1000 to 1/50000 [1]. ⋯ We demonstrate that for ~15 L of exhaled air, our device can condense reproducible volumes of EBC $({\lt} 130~ {\mu } \mathrm {L})$ in under 3 minutes (p > 0.05, n = 3). Furthermore, our results indicate that a higher concentration of glucose can be detected in the collected sample with selective valve opening (p < 0.05, n = 3). The development of this device enables a repeatable and robust collection method to enable the evaluation of correlations between analytes in EBC and blood.
-
Annu Int Conf IEEE Eng Med Biol Soc · Jul 2018
Investigation of the efficiency of the shape of chopped pulses using earthworm model.
In neural electrical stimulation, limiting the charge delivered during a stimulus pulse is essential to avoid nerve tissue damage and to save power. Previous experimental and modeling studies indicated that waveforms such as non-rectangular continuous pulses or rectangular chopped pulse were able to improve stimulation efficiency. ⋯ Results indicated that non rectangular chopped pulses activated MGF and LGF with less charge than rectangular chopped pulses. For MGF (respectively LGF), the gain of charge was up to 33.9\% (resp. 17.8\%) using chopped ramp, and up to 22.8\% (resp. 18.1\%) using chopped quarter sine.
-
Annu Int Conf IEEE Eng Med Biol Soc · Jul 2018
Deep Learning and Multi-Sensor Fusion for Glioma Classification Using Multistream 2D Convolutional Networks.
This paper addresses issues of brain tumor, glioma, grading from multi-sensor images. Different types of scanners (or sensors) like enhanced T1-MRI, T2-MRI and FLAIR, show different contrast and are sensitive to different brain tissues and fluid regions. Most existing works use 3D brain images from single sensor. ⋯ Two datasets were used for our experiments, one for classifying low/high grade gliomas, another for classifying glioma with/without 1p19q codeletion. Experiments using the proposed scheme have shown good results (with test accuracy of 90.87% for former case, and 89.39 % for the latter case). Comparisons with several existing methods have provided further support to the proposed scheme. keywords: brain tumor classification, glioma, 1p19q codeletion, glioma grading, deep learning, multi-stream convolutional neural networks, sensor fusion, T1-MR image, T2-MR image, FLAIR.